LTE uses OFDM in the downlink and DFT-spread OFDM in the uplink. FIGS. 1-3 provide an overview of LTE downlink transmissions. Referring to FIG. 1 in particular, the basic LTE physical resource can thus be seen as a time-frequency grid, where each resource element corresponds to one subcarrier during one OFDM symbol interval (on a particular antenna port).
As shown in FIG. 1, in the time domain, LTE downlink transmissions are organized into radio frames of 10 ms, each radio frame consisting of ten equally-sized subframes of 1 ms as illustrated in FIG. 2. A subframe is divided into two slots, each of 0.5 ms time duration.
As shown in FIG. 2, the resource allocation in LTE is described in terms of resource blocks, where a resource block corresponds to one slot in the time domain and 12 contiguous 15 kHz subcarriers in the frequency domain. Two in time consecutive resource blocks represent a resource block pair and correspond to the time interval upon which scheduling operates.
Transmissions in LTE are dynamically scheduled in each subframe where the base station transmits downlink assignments/uplink grants to certain UEs via the physical downlink control channel (PDCCH). The PDCCHs are transmitted in the first OFDM symbol(s) in each subframe and spans (more or less) the whole system bandwidth. A UE that has decoded a downlink assignment, carried by a PDCCH, knows which resource elements in the subframe that contain data aimed for the UE. Similarly, upon receiving an uplink grant, the UE knows which time/frequency resources it should transmit upon. In LTE downlink, data is carried by the physical downlink shared channel (PDSCH) and in the uplink the corresponding channel is referred to as the physical uplink shared channel (PUSCH). For additional information on the physical layer in LTE, see, e.g., 3GPP TS 36.213, “Physical layer procedures.”
Demodulation of sent data requires estimation of the radio channel which is done by using transmitted reference symbols (RS), i.e. symbols known by the receiver. In LTE, cell specific reference symbols (CRS) are transmitted in all downlink subframes and in addition to assist downlink channel estimation they are also used for mobility measurements performed by the UEs. LTE also supports UE specific RS aimed only for assisting channel estimation for demodulation purposes. FIG. 3 illustrates how the mapping of physical control/data channels and signals can be done on resource elements within a downlink subframe. In this example, the PDCCHs occupy the first out of three possible OFDM symbols, so in this particular case the mapping of data could start already at the second OFDM symbol. Since the CRS is common to all UEs in the cell, the transmission of CRS cannot be easily adapted to suit the needs of a particular UE. This is in contrast to UE specific RS which means that each UE has RS of its own placed in the data region of FIG. 3 as part of PDSCH.
Referring to FIG. 3, the length of the control region, which can vary on subframe basis, is conveyed in the Physical Control Format Indicator CHannel (PCFICH). The PCFICH is transmitted within control region, at locations known by terminals. After a terminal has decoded the PCFICH, it thus knows the size of the control region and in which OFDM symbol the data transmission starts.
Also transmitted in the control region is the Physical Hybrid-ARQ Indicator Channel. This channel carries ACK/NACK responses to a terminal to inform if the uplink data transmission in a previous subframe was successfully decoded by the base station or not.
Precoding
A core component in LTE is the support of MIMO antenna deployments and MIMO related techniques. LTE Rel-10 supports up to eight layer spatial multiplexing with possibly channel dependent precoding. The target is high data rates in favorable channel conditions. An illustration of spatial multiplexing is provided in FIG. 4.
As seen in FIG. 4, the information carrying symbol vector s is multiplied by an NT×r precoder matrix WNT×r, which serves to distribute the transmit energy in a subspace of the NT (corresponding to NT antenna ports) dimensional vector space. The precoder matrix is typically selected from a codebook of possible precoder matrices, and typically indicated by means of a precoder matrix indicator (PMI), which specifies a unique precoder matrix in the codebook. If the precoder matrix is confined to have orthonormal columns, then the design of the codebook of precoder matrices corresponds to a Grassmannian subspace packing problem. The r symbols in s each correspond to a layer and r is referred to as the transmission rank. In this way, spatial multiplexing is achieved since multiple symbols can be transmitted simultaneously over the same time-frequency resource element (TFRE). The number of symbols r is typically adapted to suit the current channel properties.
LTE uses OFDM in the downlink (and DFT precoded OFDM in the uplink) and hence the received NR×1 vector yn for a certain TFRE on subcarrier n (or alternatively data TFRE number n), assuming no inter-cell interference, is thus modeled byyn=HnWNT×rsn+en  equation (1)
where en is a noise vector obtained as realizations of a random process. The precoder, WNT×r, can be a wideband precoder, which is constant over frequency, or frequency-selective.
The precoder matrix is often chosen to match the characteristics of the NR×NT MIMO channel H, resulting in so-called channel dependent precoding. This is also commonly referred to as closed-loop precoding and essentially strives for focusing the transmit energy into a subspace which is strong in the sense of conveying much of the transmitted energy to the UE. In addition, the precoder matrix may also be selected to strive for orthogonalizing the channel, meaning that after proper linear equalization at the UE, the inter-layer interference is reduced.
In closed-loop precoding, the UE transmits, based on channel measurements in the forward link (downlink), recommendations to the eNodeB of a suitable precoder to use. A single precoder that is supposed to cover a large bandwidth (wideband precoding) may be fed back. It may also be beneficial to match the frequency variations of the channel and instead feedback a frequency-selective precoding report, e.g. several precoders, one per subband. This is an example of the more general case of channel state information (CSI) feedback, which also encompasses feeding back other entities than precoders to assist the eNodeB in subsequent transmissions to the UE. Such other information may include channel quality indicators (CQIs) as well as transmission rank indicator (RI).
Coordination of Nodes/Points—CoMP
In a classical cellular deployment, the intended service area is covered by several sites at different geographical positions. Each site has antennas servicing an area around the site. Often, a site is further subdivided into multiple sectors, where perhaps the most common case is to use three 120 degree wide sectors. Such a scenario is illustrated in FIG. 5. A sector corresponds to a cell and a base station associated with the cell is controlling and communicating with the UEs within that cell. The scheduling and transmissions to and reception from the UEs are to a large degree independent from one cell to another.
Simultaneous transmissions on the same frequencies will naturally interfere with each other and thus lower the quality of the reception. Interference is a major obstacle in cellular networks and in such a classical deployment scenario is primarily controlled by planning the network carefully, placing the sites at appropriate locations, tilting the antennas, etc.
Performing independent scheduling between different cells has the advantage of being simple and requiring relatively modest communication capabilities between different sites. On the other hand, the cells affect each other in that signals originating from one cell are seen as interference in nearby cells. This indicates that there are potential benefits in coordinating the transmissions from nearby cells. Frequency, time, as well as space can be exploited in the coordination to mitigate interference. Such coordination has recently received substantial interest in both academic literature and standardization of new wireless technologies. In fact, so-called coordinated multi point transmission/reception (CoMP) is considered one of the key technology components for the upcoming Release 11 of LTE (see, e.g., 3GPP TR 36.819, V1.2.0, “Coordinated Multi-Point Operation for LTE”).
The concept of a point is worth elaborating upon. A point corresponds to a set of antennas intending to cover essentially the same geographical area in essentially a similar manner. Thus a point might correspond to one of the sectors at a site, but it may also correspond to a site having one or more antennas all intending to cover a similar geographical area in a similar manner. Often, different points represent different sites. Antennas correspond to different points when they are sufficiently geographically separated and/or having antenna diagrams pointing in sufficiently different directions, but typically not when they belong to the same sector. Techniques for CoMP entail introducing dependencies in the scheduling or transmission/reception among different points, in contrast to conventional cellular systems where a point from a scheduling point of view is operated more or less independently from the other points.
Downlink CoMP can be classified into coordinated scheduling and joint transmission. In the former, the transmission to a UE comes from a single point at a time while in the latter case multiple points are simultaneously involved. FIG. 5 illustrates the use of CoMP for a group of seven points (see area enclosed by dotted line), the so-called CoMP cluster. In this particular case, each point has a one-to-one correspondence with a cell.
Obviously, coordination between sites requires some kind of communication between the sites. This can take many forms and the requirements on data rates and latency are to a large extent dependent on the exact coordination scheme being used.
Apart from the potential problem of site-to-site communication capability, coordination exploiting time and frequency is for OFDM systems like LTE easily achieved using the normal dynamic resource allocation feature which can transmit the PDSCH to a particular UE on selected RB pairs and in a certain subframe. Spatial coordination involves utilizing multiple antennas for the transmission. By modeling the signals as vector-valued signals and applying appropriate complex-valued matrix weights, the transmission can be focused in the direction (in physical space or in a more abstract vector space) of the UE while minimizing the interference to other UEs, thus increasing the SINR and ultimately the performance of the system.
Classical Versus Single Cell Deployments
The classical way of deploying a network is to let different transmission/reception points form separate cells. That is, the signals transmitted from or received at a point is associated with a cell-id that is different from the cell-id employed for other nearby points. Conventionally, each point transmits its own unique signals for broadcast (PBCH) and sync channels (PSS, SSS).
The mentioned classical strategy of one cell-id per point is depicted in FIG. 6 for a heterogeneous deployment where a number of low power (pico) points are placed within the coverage area of a higher power macro point. Note that similar principles obviously also apply to classical macro-cellular deployments where all points have similar output power and perhaps placed in a more regular fashion than what is the case for a heterogeneous deployment.
An alternative to the classical deployment strategy is to instead let all the UEs within the geographical area outlined by the coverage of the high power macro point be served with signals associated with the same cell-id. In other words, from a UE perspective, the received signals appear coming from a single cell. This is illustrated in FIG. 7. Note that only one macro point is shown, other macro points would typically use different cell-ids (corresponding to different cells) unless they are co-located at the same site (corresponding to other sectors of the macro site). In the latter case of several co-located macro points, the same cell-id may be shared across the co-located macro-points and those pico points that correspond to the union of the coverage areas of the macro points. Sync, BCH and control channels are all transmitted from the high power point while data can be transmitted to a UE also from low power points by using shared data transmissions (PDSCH) relying on UE specific RS. Such an approach has benefits for those UEs that are capable of PDSCH based on UE specific RS while UEs only supporting CRS for PDSCH (which is likely to at least include all Release 8/9 UEs for FDD) has to settle with the transmission from the high power point and thus will not benefit in the downlink from the deployment of extra low power points.
The single cell-id approach is geared towards situations in which there is fast backhaul communication between the points associated to the same cell. A typical case would be a base station serving one or more sectors on a macro level as well as having fast fiber connections to remote radio units (RRUs) playing the role of the other points sharing the same cell-id. Those RRUs could represent low power points with one or more antennas each. Another example is when all the points have a similar power class with no single point having more significance in than the others. The base station would then handle the signals from all RRUs in a similar manner.
A clear advantage of the shared cell approach compared with the classical one is that the typically involved handover procedure between cells only needs to be invoked on a macro basis. Another important advantage is that interference from CRS is greatly reduced since CRS does not have to be transmitted from every point. There is also much greater flexibility in coordination and scheduling among the points which means the network can avoid relying on the inflexible concept of semi-statically configured “low interference” subframes as in Rel-10 eICIC. A shared cell approach also allow decoupling of the downlink with the uplink so that for example path loss based reception point selection can be performed in uplink while not creating a severe interference problem for the downlink, where the UE may be served by a transmission point different from the point used in the uplink reception.
CSI-RS
As previously indicated, CRS are not the only reference symbols available in LTE. As of LTE Release-10, a new RS concept was introduced with separate UE specific RS for demodulation of PDSCH and RS for measuring the channel for the purpose of channel state information (CSI) feedback from the UE. The latter is referred to as CSI-RS. CSI-RS are not transmitted in every subframe and they are generally sparser in time and frequency than RS used for demodulation. CSI-RS transmissions may occur every 5th, 10th, 20th, 40th, or 80th subframe according to an RRC configured periodicity parameter and an RRC configured subframe offset.
A UE operating in connected mode can be requested by the base station to perform channel state information (CSI) reporting, e.g. reporting a suitable rank indicator (RI), one or more precoding matrix indices (PMIs) and a channel quality indicator (CQI). Other types of CSI are also conceivable including explicit channel feedback and interference covariance feedback. The CSI feedback assists the network in scheduling, including deciding the subframe and RBs for the transmission, which transmission scheme/precoder to use, as well as provides information on a proper user bit rate for the transmission (link adaptation). In LTE, both periodic and aperiodic CSI reporting is supported. In the case of periodic CSI reporting, the terminal reports the CSI measurements on a configured periodical time basis on the physical uplink control channel (PUCCH), whereas with aperiodic reporting the CSI feedback is transmitted on the physical uplink shared channel (PUSCH) at pre-specified time instants after receiving the CSI grant from the base station. With aperiodic CSI reports, the base station can thus request CSI reflecting downlink radio conditions in a particular subframe.
FIGS. 8a-c provide a detailed illustration of which resource elements within a resource block pair may potentially be occupied by the new UE specific RS and CSI-RS. The CSI-RS utilizes an orthogonal cover code of length two to overlay two antenna ports on two consecutive REs. As seen, many different CSI-RS patterns are available. For the case of 2 CSI-RS antenna ports we see that there are twenty different patterns within a subframe. The corresponding number of patterns is 10 and 5 for 4 and 8 CSI-RS antenna ports, respectively. For TDD, some additional CSI-RS patterns are available. A pattern may in LTE Rel-10 correspond to 1, 2, 4, or 8 CSI-RS antenna ports.
Subsequently in this disclosure, the term CSI-RS resource is used to refer to a selection of resource elements corresponding to a CSI-RS. In FIGS. 8a-c, for example, the resource elements corresponding to a CSI-RS resource share the same shading. In such a case, a resource corresponds to a particular pattern present in a particular subframe. Thus two different patterns in the same subframe or the same CSI-RS pattern but in different subframes in both cases constitute two separate CSI-RS resources. In LTE Rel-10, a CSI-RS resource can alternatively be thought of being pointed out by a combination of “resourceConfig” and “subframeConfig” which are configured by higher layers.
The CSI-RS patterns may also correspond to so-called zero-power CSI-RS, also referred to as muted REs. Zero-power CSI-RS corresponds to a CSI-RS pattern whose REs are silent, i.e., there is no transmitted signal on those REs. Such silent patterns are configured with a resolution corresponding to the four antenna port CSI-RS patterns. Hence, the smallest unit to silence corresponds to four REs.
The purpose of zero-power CSI-RS is to raise the SINR for CSI-RS in a cell by configuring zero-power CSI-RS in interfering cells so that the REs otherwise causing the interference are silent. Thus, a CSI-RS pattern in a certain cell is matched with a corresponding zero-power CSI-RS pattern in interfering cells. Raising the SINR level for CSI-RS measurements is particularly important in applications such as coordinated multi point (CoMP) or in heterogeneous deployments. In CoMP, the UE is likely to need to measure the channel from non-serving points and interference from the much stronger serving point would in that case be devastating. Zero-power CSI-RS is also needed in heterogeneous deployments where zero-power CSI-RS in the macro-layer is configured so that it coincides with CSI-RS transmissions in the pico-layer. This avoids strong interference from macro nodes when UEs measure the channel to a pico node.
The PDSCH is mapped around the REs occupied by CSI-RS and zero-power CSI-RS so it is important that both the network and the UE are assuming the same CSI-RS/zero power CSI-RS configuration or else the UE is unable to decode the PDSCH in subframes containing CSI-RS or their zero-power counterparts.
CSI Feedback for CoMP
To assist scheduling and link adaptation when performing CoMP, it is useful to let the UE feedback CSI corresponding to the channels of multiple points to the network. Such feedback allows the network to assess the impact on system performance (i.e., taking multiple points into account) of scheduling a UE on a certain resource and with a certain precoder. This may then be exploited for devising efficient scheduling strategies across multiple points.
CSI feedback for CoMP can come in many different forms but a common scheme is to let each UE report CSI feedback for each CSI-RS resource in a set of relevant CSI-RS resources that are used for the feedback reporting, the so-called (CoMP) reporting set. A relevant CSI-RS resource typically corresponds to the transmission of a CSI-RS pattern that can be heard sufficiently well by the UE. Often, such a transmission would be conducted from a specific point, meaning that per CSI-RS resource feedback can be thought of as CSI feedback per point.
FIG. 9 illustrates an example prior art CSI reporting configuration, in which a CSI report 20 transmitted by a UE includes separately-determined feedback for each CSI-RS resource (i.e., CSI feedback per CSI-RS resource), and each of a plurality of precoders uses a separate one of the feedback values. Also, in FIG. 9, CQI and precoder Wk (implied from PMI and RI) are determined separately for each CSI-RS resource. Thus, CQI, rank, and precoder would be determined separately for each CSI-RS resource that is reported. A similar concept is already adopted for carrier aggregation where CSI for each carrier (cell) is determined separately. Transmission formats and procedures for carrier aggregation can thus be reused for CoMP feedback, thus greatly simplifying the introduction of new feedback into the specifications. Per CSI-RS resource feedback also has the benefit of limiting UE complexity in that separately determining CSI for each CSI-RS resource is less complex than jointly determining CSI for all CSI-RS resources at once.
Problems with Existing Solutions
MIMO with spatial multiplexing based on rank adaptation is commonly employed in LTE to match the transmission to the properties of the channel, thereby improving the performance and offering high peak rates under good channel conditions. Existing solutions, however, do not clearly contemplate how to perform efficient rank determination for CoMP CSI feedback.